Critique 209: Does alcohol intake increase the risk of obesity? — 13 December 2017
Downer MK, Bertoia ML, Mukamal KJ, Rimm EB, Stampfer MJ. Change in Alcohol Intake in Relation to Weight Change in a Cohort of US Men with 24 Years of Follow-Up. Obesity 2017;25:1988-1996. doi:10.1002/oby.21979
Objective: The aim of this study was to prospectively investigate the potential effects of alcohol by subtype on reported long-term weight change.
Methods: This study examined changes in alcohol intake (total, wine, light beer, regular beer, and liquor) and simultaneous changes in reported body weight within 4-year periods from 1986 to 2010 from US men in the Health Professionals Follow-Up Study. The study adjusted for age, changes in lifestyle and dietary covariates, and cardiovascular risk factors.
Results: The study included observations of 44,603 four-year periods from 14,971 men. Total alcohol, total beer, regular beer, and liquor intakes, modeled as the increase in weight per increase in drinks per day, were each directly associated with moderate weight gain over the 4-year periods, in pounds: total alcohol: 0.23 (0.10 to 0.35); total beer: 0.29 (0.08 to 0.51); regular beer: 0.61 (0.22 to 1.00); and liquor: 0.28 (0.09 to 0.48). Results for wine and light beer were as follows: wine: 0.16 (- 0.04 to 0.36) and light beer: – 0.38 (- 1.07 to 0.08). Results were strongest for men <55 years old.
Conclusions: Increased alcohol consumption was associated with minor reported weight gain at levels unlikely to be clinically meaningful. Beverage-specific differences were not substantial enough to make dietary recommendations for weight loss or maintenance by beverage type. The greatest risk of weight gain was among men who increased consumption to levels well above moderation.
The public, and scientists, have long been concerned about the relation between alcohol consumption and weight gain. A frequent comment, especially among women, is “I would like to have a glass of wine, but I am worried about the calories.” Indeed, all alcoholic beverages contain calories, but it may be somewhat surprising that most epidemiologic studies do not find that light or moderate drinkers weigh more than their abstaining peers (Sayon-Orea et al), and some even show lower weight among moderate drinkers than among abstainers (Thomson et al; Dumesnil et al; MacInnis et al).
The potential lack of increased risk of obesity among consumers of alcohol has been a scientific question for many years: could it be a lower effect of alcohol calories in relation to those from other foods, a decrease in other calories among drinkers, changes in satiety from alcohol consumption, or other factors? The physiologic mechanisms for the association remain unclear.
The present study is important as it describes the effects on weight of changes in alcohol intake over repeated 4-year periods among almost 15,000 health professionals, followed over 24 years. Its strengths include being a cohort study with repeated assessments of alcohol intake and body weight; the multiple assessments served indirectly as adjustments for within-person variation and reporting error. Further, as the investigators state, “The relative demographic homogeneity of the population is a potential advantage as it helps reduce residual confounding by by unmeasured lifestyle characteristics associated with alcohol intake patterns.”
Without any consideration of alcohol consumption, the analyses indicate an average increase in weight of the subjects of 2 pounds for each 4-year period of observation. Overall, increasing alcohol intake was associated with a slightly greater average weight gain over the 4-year periods: 0.23 pounds for total alcohol intake; 0.61 pounds for regular beer; and 0.28 pounds for liquor. Wine intake and light beer intake were not significantly related to weight change.
As for total alcohol intake, in comparison with subjects who did not change their intake over time, those increasing their intake by 2 or more drinks/day gained an average of 0.62 pounds while those who decreased their alcohol intake by 1 to ≤ 2 drinks/day showed a decrease in weight of 0.55 pounds (but those who decreased their alcohol intake more did not change their weight). It should be noted that, in general, changes in alcohol intake related to weight changes in the expected direction (a slight increase for subjects increasing alcohol intake and a small decrease for most groups decreasing their alcohol intake), but all of the differences were very small.
The investigators conclude that any effects of drinking on weight are very minor, and probably of no clinical significance. While effects on weight gain were slightly lower among consumers of wine and light beer than for those consuming regular beer or spirits, they were not large. The study does not present data on the mechanisms of such associations.
Specific comments from Forum members: Forum member Finkel noted: The relationship between weight and drinking used to engage my interest, much to my puzzlement. I recall odd and mysterious results in a number of studies, including unexpected differences between the sexes and for wine versus other beverages. I haven’t seen much on this area lately, so am pleased to welcome this study, especially because it is from a group with a reliable track record. The results reported by this paper are not out of line with those of the previous with which I am familiar. Basically, the bottom line message is ‘no big deal.’ The question, though, remains whether alcohol or some element in wine alters caloric metabolism. Or is it just that drinking may variably alter food intake – or is this too simple?”
Finkel continued: “There still is misconception, even among physicians, that dry wine, for example, is ‘full of carbohydrates,’ and thus contraindicated even for mild diabetics, and that even moderate drinking leads to significant weight gain. This paper might help dispel that among those willing to listen.” Added reviewer Ellison: “Given that patients with diabetes who are moderate drinkers have a considerable reduction in their risk of cardiovascular disease (the leading cause of death among diabetics), it is important that they be given accurate information on alcohol effects.”
Reviewer van Velden commented: “In our own studies, we have found that moderate wine consumption had a beneficial effect on patients with the Metabolic Syndrome. The refined carbohydrates in the grape juice have been fermented to ethanol, and do not affect the blood glucose concentration and insulin secretion. Many patients have insulin resistance and cannot tolerate glucose or fructose in the diet. Our wine=drinking patients did not gain weight, but this may also be because they have a balanced, Mediterranean-type diet.
“It seems that there is a world-wide reaction against the old food guide pyramid, as proposed by the US dietary guidelines, with the emphasis on carbohydrates and low fat intake. Any diet that limits the intake of refined carbohydrate is beneficial. Our studies do not suggest that moderate wine intake leads to weight gain.”
Reviewer Goldfinger wrote: “In general wine reportedly has about 2 – 4 gm CHO/80 – 120 kcal per glass depending on size and color. There are clearly misconceptions with respect to CHO load from wine. Some people just don’t get that the sugars are gone after fermentation (minimal residual sugars are just that).
“As a practitioner, I can say that patients report weight loss when wine and other alcoholic beverages are excluded. My suspicion is that those who have weight shifts based on moderation of their alcohol consumption are consuming greater volumes. A sensitive and scientific review of this paper and this topic is warranted, linking this topic with salutary effects on metabolic syndrome seen previously.”
Forum member Thelle noted: “This is obviously an area which deserves further research. One of the most recent reviews on the issue (Traversy et al) concludes: ‘”In general, recent prospective studies show that light-to-moderate alcohol intake is not associated with adiposity gain while heavy drinking is more consistently related to weight gain. Experimental evidence is also mixed and suggests that moderate intake of alcohol does not lead to weight gain over short follow-up periods. There are few experiments, but one comparing water and wine is by Golan et al, which concludes: ‘Moderate wine consumption, as part of a Mediterranean diet, in persons with controlled diabetes did not promote weight gain or abdominal adiposity.’”
Reviewer Ursini commented: “Let me put the question from the point of view of biochemistry and metabolism (textbook level): ethanol is practically fat. It produces AcCoA from which you get carbonic anhydride or triglycerides. Glucose and insulin make the difference. Thus it would be reasonable to expect that it is the CHO content in the diet together with ethanol that favours obesity. Ethanol just provides the carbon atoms that will become fat.”
Alcohol and body weight: an association affected by many factors: Forum member Skovenborg provided a good overview of how a number of factors may modify the association between alcohol consumption and body weight. “Contrary to the conventional perspective, recent research has shown that calories from different foods affect hunger, hormones, and energy expenditure in ways that cannot be explained by consideration of caloric balance alone. For example, energy expenditure decreased by 325 kcal per day among volunteers in a crossover study who consumed a low-fat diet compared with when they consumed a calorie-matched low-carbohydrate diet (Ebbeling et al).
“Genetic factors must also be considered in studies of the effect of alcohol intake on body weight. A study of the response to long-term overfeeding in identical twins found a mean weight gain of 8,1 kg, but the range was 4,3 to 13,3 kg. The alterations in body weight were characterized by about three times more variance between the pairs of twins than within pairs. The man who gained the most weight (13.3 kg) had no evidence of energy dissipation by any mechanism whereas in the man who gained the least weight (4.3 kg) only about 40 percent of the extra calories were deposited as body tissues (Bouchard et al).
“A subject’s BMI may also interact with alcohol consumption in the effects on body weight. In an experimental study of the effect of adding 630 calories of alcohol to the diet of 12 men, the addition of alcohol did not cause weight gain in lean individuals; however half of the obese individuals definitely gained weight when the diet was supplemented with alcohol (Crouse et al). In the present study, Downer et al examined the possible effect modification of baseline body weight by stratifying on BMI (but only for BMI >25 and BMI 25-27,5) and found no difference.
“The drinking patterns of subjects may also relate to the effects of alcohol intake on body weight. In a large Danish cross-sectional study of 25.325 men and 24.552 women aged 50-65 years, for a given level of total alcohol intake, obesity was inversely associated with drinking frequency, whereas the amount of alcohol intake was positively associated with obesity. These results indicate that frequent drinking of small amounts of alcohol is the optimal drinking pattern in this relation (Tolstrup et al). A French cross-sectional study of 7,855 men aged 50–59 years found the number of drinking episodes inversely correlated with body mass index and waist circumference. The odds ratio for obesity was 1.8 (CI 1.3–2.4) for occasional (1–2 days/week) and 1.6 (1.2–2.1) for more frequent (3–5 days/week, but not daily) drinkers compared with daily drinkers (Dumesnil et al). In the present study, Downer et al did not report the drinking patterns of the cohort; however, a good guess would be that most Health Professionals were not frequent drinkers. Apart from the missing information on drinking patterns the authors must be complimented on their heroic efforts to adjust for a multitude of confounders.”
Reviewer Waterhouse described the carefully controlled intervention trial done by Rumpler et al on the effects of alcohol consumption on weight. “Rumpler et al at USDA had a fairly definitive conclusion on energy utilization from alcohol consumption, based on a trial comparing the consumption of carbohydrate vs ethanol in a very controlled environment. Weight gain over 16 weeks was very slight and was negative in a low-fat diet, and if anything ethanol in the diet instead of carbohydrate reduced final weight. These scientists concluded: ‘Total energy expenditure for 7 weeks was the same when subjects consumed either ethanol or carbohydrate. These data clearly show that, on an energy basis, ethanol and carbohydrate are utilized in the diet with the same efficiency. These data are consistent with the efficiency of use of alcohol for maintenance of metabolizable energy [with that from alcohol] being the same as that for carbohydrate.’”
Bouchard C et al. The response to long-term overfeeding in identical twins. N Engl J Med 1990;322:1477-1482.
Crouse JR, Grundy SM. Effects of alcohol on plasma lipoproteins and cholesterol and triglyceride metabolism in man. J Lipid Res 1984;25:486-496.
Dumesnil C, Dauchet L, Ruidavets JB, et al. Alcohol consumption patterns and body weight. Ann Nutr Metab 2013;62:91-97.
Ebbeling CB, Swain JF, Feldman HA, et al. Effects of dietary composition on energy expenditure during weight-loss maintenance. JAMA 2012;307:2627-2634.
Golan R et al. Effects of initiating moderate wine intake on abdominal adipose tissue in adults with type 2 diabetes: a 2-year randomized controlled trial. Public Health Nutr 2017;20:549-555. doi: 10.1017/S1368980016002597.
MacInnis RJ, Hodge AM, Dixon HG, et al. Predictors of increased body weight and waist circumference for middle-aged adults. Public Health Nutr 2014;17:1087-1097.
Rumpler WV, Rhodes DG, Baer DJ, Conway JM, Seale JL. Energy value of moderate alcohol consumption by humans. Am J Clin Nutr 1996;64:108-114.
Sayon-Orea C, Bes-Rastrollo M, Nunez-Cordoba JM, Basterra-Gortari FJ, Beunza JJ, Martinez-Gonzalez MA. Type of alcoholic beverage and incidence of overweight/obesity in a Mediterranean cohort: the SUN project. Nutrition 2011;27:802-808.
Thomson CA, Wertheim BC, Hingle M, et al. Alcohol consumption and body weight change in postmenopausal women: results from the Women’s Health Initiative. Int J Obes (Lond) 2012;36:1158-1164.
Tolstrup JS et al. The relation between drinking pattern and body mass index and waist and hip circumference. Int J Obes 2005;29:490-497.
Traversy G, Chaput JP. Alcohol Consumption and Obesity: An Update. Curr Obes Rep 2015;4:122-130. doi: 10.1007/s13679-014-0129-4.
The public, and scientists, have long been concerned about the relation between alcohol consumption and weight gain, as all alcoholic beverages contain calories. However, most epidemiologic studies do not find that light or moderate drinkers weigh more than their abstaining peers, and some even show lower weight among moderate drinkers than among abstainers. The reasons for this apparent effect are unclear.
The present study is important as it describes the effects on weight of changes in alcohol intake over repeated 4-year periods among a large cohort of health professionals, followed for 24 years. They found very slight increases in weight (mostly less than one-half pound over a four-year period) for subjects consuming alcoholic beverages; these were statistically significant for regular beer and liquor, but not for wine or light-beer. Overall, the largest increases in weight (still, only an increase of 0.6 pounds) were seen for subjects increasing their alcohol by an average of 2 or more drinks/day. For subjects decreasing their alcohol intake over the periods, there was a slight decrease in weight, still only about one-half pound or less.
The investigators conclude that any effects of drinking on weight are very minor, and probably of no clinical significance. While effects on weight gain were slightly lower among consumers of wine and light beer than for those consuming regular beer or spirits, the differences were not large. The study does not present data on the mechanisms of such associations.
Forum members considered this to be a very well-done analysis of repeatedly collected exposure and outcome data among health professionals over an extended period of time. Adjustments were made for known potential confounders associated with changes in weight and appropriate sensitivity analyses were done. While the paper indicates that calories from alcohol are metabolized similarly to those from other foods and do affect weight change, the changes associated with moderate alcohol intake appear to be very minor and would be expected to have very little effect on the development of obesity.
* * * * * *
Contributions to this critique by the International Scientific Forum on Alcohol Research have been provided by the following members:
R. Curtis Ellison, MD, Professor of Medicine, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, University of Arizona School of Medicine, Tucson, AZ, USA
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Dag S. Thelle, MD, PhD, Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, Norway; Section for Epidemiology and Social Medicine, Sahlgrenska Academy, University of Gothenburg, Sweden
Fulvio Ursini, MD, Dept. of Biological Chemistry, University of Padova, Padova, Italy
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA